[Home][2015 Index] Re: [TCML] Variable Inductive Ballast [Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [TCML] Variable Inductive Ballast


Many years ago, I bought a medium sized TC from Harry Goldman. He designed it with two esoteric unobtanium, epoxy potted, internally current limited 10 kVAC transformers in parallel that apparently came from some sort of radar installation. One of them failed with the secondary shorted.

As a replacement, I had a large Hammond 9 KV, 500 mA potential transformer, which had to be current limited to avoid overloading the TC. On eBay, I found a 2 KVA saturable reactor (SR) for a decent price.

The SR is wired in series with the primary of the potential transformer. It has three windings on a figure "8" core, connected internally and 4 externally accessible terminals. Two terminals are the DC control winding, and two are for the controlled AC line current.

You feed the control winding with 0 - 75 volts of full wave rectified, but unfiltered DC, from a small Variac, at less than one amp. The SR passes only a fraction of an amp through the high current winding when the DC control voltage is turned off, and the impedance of the SR decreases as the DC control voltage increases, up to the maximum current capacity of the unit.

When driven with the SR / potential transformer combo, the TC runs better (much smoother firing, longer sparks), than it ever had with a Variac and the original power transformers.

Interestingly, there is a little bit of delayed response of the SR. If you flip off the DC control voltage, the TC discharge tapers off gradually over about 1 - 1.5 seconds, rather than stopping instantly.

There was a great deal of research in this field before WWII about magnetic amplifiers, and there are many web pages about them. These were very effective in the days before solid state controls were available. To me, the big advantage of them in TC use is that they are much less likely to release their "magic smoke" than today's IGBT circuits.

I read somewhere a long time ago that you could make a home built SR by connecting two matched MOTs with their primaries in series, and their HV secondaries in opposing series. You would apply the control DC to the nominal HV secondary windings. If the MOTs are closely matched, the opposed secondary windings result in a net zero potential despite the voltages on each winding induced by the primary current.

Only disadvantage is that the two MOTs in series are only good for controlling 120 VAC. I've never had time to try this arrangement out, but the principle sounded right. Aside from that, I've never tried to create my own SRs. Wish I had the time to try, but that's a scarce commodity around here.


On 9/30/2015 6:27 PM, Phil wrote:
Any tips, design details, or weblinks that you may have found useful when
building your saturable ballasts?

Phil Tuck

Tesla mailing list